DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant’s arguments, see page 8, filed 08/18/2025, with respect to claim rejections under 35 U.S.C. 112 have been fully considered and are persuasive. The rejections of Claims 4, 11, 16, 17, and 20 under 35 U.S.C. 112 have been withdrawn.
Applicant's arguments filed 08/18/2025, with respect to claim rejections under 35 U.S.C. 101 have been fully considered but they are not persuasive.
Beginning on page 11, Applicant states that “the claims [1 and 19] are directed to a practical application through a structured computing system” reciting the physical components. This is not persuasive because the amended limitations merely recite a computer to perform in its expected capacity (see MPEP 2106.05(f)(2) Whether the claim invokes computers or other machinery merely as a tool to perform an existing process. Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.). Under broadest reasonable interpretation (BRI), the claim recites a computer collecting weather data, organizing that data, analyzing the data through a model, and transmitting/providing the output of the model. Predicting a specific type of weather condition is not significantly more and making the output available for future decisions is considered intended use.
Claim 8 remains rejected for similar reasoning.
Applicant's arguments filed 08/18/2025, with respect to claim rejections under 35 U.S.C. 102 have been fully considered but they are not persuasive.
On page 13, Applicant remarks that the cited portions of the Moon reference do not discloses “a unit that outputs frost prediction for the next day being calculated by a frost learning model based on generated frost prediction training data”. This is not persuasive because continuously collects data and trains a model to predict frost the next morning (Moon p. 2, Col. 1: para. 4, a) Evaluation: […] We predict the possibility of frost in the next morning using microclimate data. We use machine learning toolkits available in GraphLab [8] to train and evaluate five machine learning algorithms. We use 80% data for training and the remaining 20% for testing. Also see p. 2, Col. 2: para. 1, b) User Interface: […] We forecast at 11 PM through the web and mobile services so that farmers can proactively implement preventive actions. For farmers who receive forecast services automatically, they are notified with updated, more accurate information at 1 AM. and Section III. The proposed platform is successfully deployed in 12 locations and is continuously collecting microclimate data.).
The remarks that “Moon does not teach the intermediary transformation of the data into training data” and “does not apply that transformed dataset to a dedicated model to generate the next day inferences” is not persuasive for similar reasons in that Moon continuously collects microclimate data in the regions to train and evaluate models to use for prediction, where some of the data is used for training and the rest for testing and prediction.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Specifically, representative Claim 1 recites:
A frost prediction system comprising:
a user terminal;
a computer system; and
a weather observation data collection sensor attached to a meteorological station and configured to collect weather observation data and to transmit real-time weather observation data to an input interface of the computer system;
wherein the computer system includes:
the input interface;
an output interface;
one or more processors; and
memory including instructions that, when executed by the one or more processors result in:
a training data generation unit configured to generate frost prediction training data by using the collected weather observation data; and
a frost prediction unit configured to perform frost prediction for the next day by applying the generated frost prediction training data to a frost learning model; and
a frost prediction information providing unit configured to output, through the output interface to the user terminal, frost prediction information for the next day, the frost prediction information being calculated by the frost learning model based on the generated frost prediction training data.
The claim limitations in the abstract idea have been highlighted in bold above; the remaining limitations are “additional elements”.
Under the Step 1 of the eligibility analysis, we determine whether the claims are to a statutory category by considering whether the claimed subject matter falls within the four statutory categories of patentable subject matter identified by 35 U.S.C. 101: Process, machine, manufacture, or composition of matter. The above claim is considered to be in a statutory category (machine).
Under the Step 2A, Prong One, we consider whether the claim recites a judicial exception (abstract idea). In the above claim, the highlighted portion constitutes an abstract idea because, under a broadest reasonable interpretation, it recites limitations that fall into/recite an abstract idea exceptions. Specifically, under the 2019 Revised Patent Subject matter Eligibility Guidance, it falls into the grouping of subject matter when recited as such in a claim limitation, that covers mathematical concepts (mathematical relationships, mathematical formulas or equations, mathematical calculations) and mental processes – concepts performed in the human mind including an observation, evaluation, judgement, and/or opinion.
For example, the step of “perform frost prediction for the next day by applying the generated frost prediction training data to a frost learning model and the frost prediction information being calculated by the frost learning model based on the generated frost prediction training data (input data into a model to obtain output, i.e. regression)” is treated by the Examiner as belonging to mathematical concept grouping, while the steps of “to generate frost prediction training data by using the collected weather observation data (deciding which data to include as training data); and
output frost prediction information for the next day (communicating results)” are treated as belonging to mental process grouping.
Similar limitations comprise the abstract ideas of Claims 8 and 19.
Next, under the Step 2A, Prong Two, we consider whether the claim that recites a judicial exception is integrated into a practical application.
In this step, we evaluate whether the claim recites additional elements that integrate the exception into a practical application of that exception.
The above claims comprise the following additional elements:
Claim 1: A frost prediction system comprising: a user terminal; a computer system; and a weather observation data collection sensor attached to a meteorological station and configured to collect weather observation data and to transmit real-time weather observation data to a server; wherein the computer system includes: the input interface; an output interface; one or more processors; and memory including instructions; a training data generation unit; a frost prediction unit; a frost prediction information providing unit
Claim 8: A frost prediction method comprising: transmitting, by weather observation data collection sensors attached to each of a plurality of meteorological stations and configured to collect weather observation data, real-time weather observation data to a server; receiving by an input interface of a computer system, the weather observation data;
Claim 19: A frost prediction model learning method comprising: receiving at an input interface of a computer system, weather observation data; receiving, at the input interface of the computer system, date information corresponding to a frosty day.
The additional element in the preamble of “A frost prediction system/model/model learning method” is not qualified for a meaningful limitation because it only generally links the use of the judicial exception to a particular technological environment or field of use. A weather observation data collection sensor attached to a meteorological station and configured to collect weather observation data and to transmit real-time weather observation data to a server, transmitting, by various kinds of weather observation data collection sensors attached to a meteorological station and configured to collect weather observation data, real-time weather observation data to a server, or obtaining date information of a frosty day, receiving by an input interface of a computer system, the weather observation data, receiving at an input interface of a computer system, weather observation data, and receiving, at the input interface of the computer system, date information corresponding to a frosty day represent mere data gathering steps and only adds an insignificant extra-solution activity to the judicial exception. A memory or server (generic memory), a computing system with input/output interfaces and one or more processors, a training data generation unit, and a frost prediction unit (generic processors) are generally recited and are not qualified as particular machines. Additionally, a computer to perform in its expected capacity is considered mere instructions to apply an exception (see MPEP 2106.05(f)(2) Whether the claim invokes computers or other machinery merely as a tool to perform an existing process. Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.)
In conclusion, the above additional elements, considered individually and in combination with the other claim elements do not reflect an improvement to other technology or technical field, and, therefore, do not integrate the judicial exception into a practical application. Therefore, the claims are directed to a judicial exception and require further analysis under the Step 2B.
However, as discussed in the previous office action, the above claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception (Step 2B analysis).
The claims, therefore, are not patent eligible.
With regards to the dependent claims, claims 2-4, 6, 7, 9-18, and 20 provide additional features/steps which are part of an expanded algorithm, so these limitations should be considered part of an expanded abstract idea of the independent claims.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 1, 2, 8, 9, 12, and 19 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Moon et al. ("Microclimate-Based Predictive Weather Station Platform: A Case Study for Frost Forecast." 2017 IEEE High Performance Extreme Computing Conference.), hereinafter “Moon”.
Regarding Claim 1, Moon teaches a frost prediction system comprising:
a user terminal (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions.);
a computer system (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. We forecast at 11 PM through the web and mobile services so that farmers can proactively implement preventive actions.); and
a weather observation data collection sensor attached to a meteorological station and configured to collect weather observation data and to transmit real-time weather observation data to an input interface of the computer system (Moon p. 1, Section II, para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. See Fig. 1 Sensor Node Kit and Web Data and Data Collection);
wherein the computer system includes:
the input interface (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. […] Moreover, it provides an interface for farmers to easily provide the system feedback for more accurate data collection.);
an output interface (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions.);
one or more processors (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. Note one or more processors is inherent to the disclosed platform); and
memory (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. Note memory is inherent to the disclosed platform) including instructions that, when executed by the one or more processors result in:
a training data generation unit configured to generate frost prediction training data by using the collected weather observation data (Moon p. 2, Section II. a) para. 1, We use machine learning toolkits available in GraphLab [8] to train and evaluate five machine learning algorithms. We use 80% data for training and the remaining 20% for testing.); and
a frost prediction unit configured to perform frost prediction for the next day by applying the generated frost prediction training data to a frost learning model (Moon p. 2, Section II. a) para. 1, We predict the possibility of frost in the next morning using microclimate data. […] Based on our analysis, the proposed system can inform the possibilities of frost to farmers in advance such that they can proactively take preventive actions to protect the crops from frost damage. See Fig. 1, Data Processing prediction); and
a frost prediction information providing unit configured to output, through the output interface to the user terminal, frost prediction information for the next day, the frost prediction information being calculated by the frost learning model based on the generated frost prediction training data (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. We forecast at 11 PM through the web and mobile services so that farmers can proactively implement preventive actions. For farmers who receive forecast services automatically, they are notified with updated, more accurate information at 1 AM. The location of observation stations are displayed on the map, and frost prediction/occurrence information, micro-weather information, etc., are displayed in real time on our project website1.).
Regarding Claim 2, Moon further teaches a database in which date information of a frosty day and the weather observation data collected by the weather observation data collection sensor are stored (Moon p. 1 Section II, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. Solar radiation is calculated as cumulative light intensity per day. As shown in several recent studies, temperature inversion plays an important role in determining the formation of precipitation. As a result, convection produced by the heating of air from below is limited to levels below the inversion, and then frost is highly likely to occur. See Fig. 1 Data Server),
wherein the frost learning model performs learning by matching the frost prediction training data, generated by using the weather observation data stored in the database, with the date information of the frosty day (Moon p. 2, Col. 1: para. 3, After collecting all these information and calculating ∆T, we performed data analysis on data patterns on days when frost occurred by using five machine learning (ML) algorithms. Also see p. 2, Section II. a) para. 1, Evaluation: For evaluation, we collected frost data in four regions of Yeoungcheon, South Korea, from October 1 to November 23 in 2015. The number of actual frost occurrence is 19 out of entire 216 observed data points (54 days per each station).).
Regarding Claim 8, Moon teaches a frost prediction method comprising:
transmitting, by weather observation data collection sensors attached to each of a plurality of meteorological stations and configured to collect weather observation data, real-time weather observation data to a server (Moon p. 1, Section II, para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. See Fig. 1 Sensor Node Kit and Web Data and Data Collection);
receiving, by an input interface of a computer system, the weather observation data (Moon p. 1, Section II, para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. See Fig. 1 Sensor Node Kit and Web Data and Data Collection);
generating, by the computer, frost prediction training data by using the weather observation data (Moon p. 2, Section II. a) para. 1, We use machine learning toolkits available in GraphLab [8] to train and evaluate five machine learning algorithms. We use 80% data for training and the remaining 20% for testing.); and
applying, by the computer system, the frost prediction training data to a frost learning model to predict a probability of frost occurrence for a subsequent day (Moon p. 2, Section II. a) para. 1, We predict the possibility of frost in the next morning using microclimate data. […] Based on our analysis, the proposed system can inform the possibilities of frost to farmers in advance such that they can proactively take preventive actions to protect the crops from frost damage. See Fig. 1, Data Processing prediction); and
transmitting, by an output interface of the computer system, frost prediction information to a user terminal based on the prediction (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. We forecast at 11 PM through the web and mobile services so that farmers can proactively implement preventive actions. For farmers who receive forecast services automatically, they are notified with updated, more accurate information at 1 AM. The location of observation stations are displayed on the map, and frost prediction/occurrence information, micro-weather information, etc., are displayed in real time on our project website1.).
Regarding Claim 9, Moon further teaches wherein the frost learning model performs learning by matching the frost prediction training data, generated by using the weather observation data stored in a database in which date information of a frosty day and the weather observation data collected by the weather observation data collection sensors are stored (Moon p. 1 Section II, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. Solar radiation is calculated as cumulative light intensity per day. As shown in several recent studies, temperature inversion plays an important role in determining the formation of precipitation. As a result, convection produced by the heating of air from below is limited to levels below the inversion, and then frost is highly likely to occur. See Fig. 1 Data Server), with date information of the frosty day (Moon p. 2, Col. 1: para. 3, After collecting all these information and calculating ∆T, we performed data analysis on data patterns on days when frost occurred by using five machine learning (ML) algorithms. Also see p. 2, Section II. a) para. 1, Evaluation: For evaluation, we collected frost data in four regions of Yeoungcheon, South Korea, from October 1 to November 23 in 2015. The number of actual frost occurrence is 19 out of entire 216 observed data points (54 days per each station).).
Regarding Claim 12, Moon further teaches transferring the calculated frost prediction information to a user terminal (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. We forecast at 11 PM through the web and mobile services so that farmers can proactively implement preventive actions. For farmers who receive forecast services automatically, they are notified with updated, more accurate information at 1 AM. Moreover, it provides an interface for farmers to easily provide the system feedback for more accurate data collection. The location of observation stations are displayed on the map, and frost prediction/occurrence information, micro-weather information, etc., are displayed in real time on our project website1).
Regarding Claim 19, Moon teaches a frost prediction model learning method comprising:
receiving, by an input interface of a computer system, the weather observation data (Moon p. 1, Section II, para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. See Fig. 1 Sensor Node Kit and Web Data and Data Collection);
storing, in a database of the computer system, the received weather observation data (Moon p. 1, Section II, para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. See Fig. 1 Sensor Node Kit and Web Data and Data Collection. The data must inherently be stored for transmission and analysis);
receiving, at the input interface of the computer system, data information corresponding to a frosty day (Moon p. 2, Section I. a) Evaluation: For evaluation, we collected frost data in four regions of Yeoungcheon, South Korea, from October 1 to November 23 in 2015. The number of actual frost occurrence is 19 out of entire 216 observed data points (54 days per each station). The data is recorded and must have some time information in order to be properly analyzed).
Generating, by the computer system, frost prediction training data by using the weather observation data and the date information (Moon p. 2, Section II. a) para. 1, We use machine learning toolkits available in GraphLab [8] to train and evaluate five machine learning algorithms. We use 80% data for training and the remaining 20% for testing.);
obtaining date information of a frosty day (Moon p. 1, Section II, para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. See Fig. 1 Sensor Node Kit and Web Data and Data Collection); and
applying, by the computer system, the frost prediction training data for a day before to the frost day to a frost prediction model to predict a probability of frost occurrence (Moon p. 2, Section II. a) para. 1, We predict the possibility of frost in the next morning using microclimate data. […] Based on our analysis, the proposed system can inform the possibilities of frost to farmers in advance such that they can proactively take preventive actions to protect the crops from frost damage. See Fig. 1, Data Processing prediction); and
transmitting, by an output interface of the computer system, frost prediction information to a user terminal based on the prediction (Moon p. 2, Section II. b) User Interface: The proposed platform provides both the web and a mobile services for farmers who can subscribe agricultural services for their farming decisions. We forecast at 11 PM through the web and mobile services so that farmers can proactively implement preventive actions. For farmers who receive forecast services automatically, they are notified with updated, more accurate information at 1 AM. The location of observation stations are displayed on the map, and frost prediction/occurrence information, micro-weather information, etc., are displayed in real time on our project website1.).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 3, 10, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Moon (as stated above).
Regarding Claim 3, Moon is not relied upon to explicitly teach all of wherein the weather observation data include a temperature, humidity, a grass temperature, a wind speed, a soil temperature, an amount of precipitation, and insolation.
However, Moon further teaches through prior wherein the weather observation data include a temperature, humidity, a grass temperature, a wind speed, a soil temperature, an amount of precipitation, and insolation (Moon p. 1, Col. 2: para. 2, To forecast frost more accurately, several prior studies used microclimate data such as temperatures measured at lower altitude, grass minimum temperature, diurnal, average relative humidity, minimum relative humidity, mean wind speed, etc [5], [6]. The grass temperature is measured using thermometers just above the grass, about 10cm above ground. In [7], the authors used the cloud cover, the atmospheric temperature measured at midnight, and 5-day precipitation and predicted the possibility of frost with 87% of accuracy. Also see Section I. para. 4, To forecast frost more accurately, several prior studies used microclimate data such as temperatures measured at lower altitude, grass minimum temperature, diurnal, average relative humidity, minimum relative humidity, mean wind speed, etc [5], [6]. The grass temperature is measured using thermometers just above the grass, about 10cm above ground. In [7], the authors used the cloud cover, the atmospheric temperature measured at midnight, and 5-day precipitation and predicted the possibility of frost with 87% of accuracy. And Section II. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. Solar radiation is calculated as cumulative light intensity per day. See Fig. 1. Sensor Node Kit).
It would have been obvious to one of ordinary skill in the art prior to the effective filing date of the instant application, to modify Moon to explicitly teach wherein the weather observation data include a temperature, humidity, a grass temperature, a wind speed, a soil temperature, an amount of precipitation, and insolation, because several prior studies have used the listed observation data to improve prediction accuracy (Moon p. 1, Col. 2: para. 2).
Regarding Claim 10, Moon is not relied upon to explicitly teach all of wherein the weather observation data include a temperature, humidity, a grass temperature, a wind speed, a soil temperature, an amount of precipitation, and insolation.
However, Moon further teaches wherein the weather observation data include a temperature, humidity, a grass temperature, a wind speed, a soil temperature, an amount of precipitation, and insolation (Moon p. 1, Col. 2: para. 2, To forecast frost more accurately, several prior studies used microclimate data such as temperatures measured at lower altitude, grass minimum temperature, diurnal, average relative humidity, minimum relative humidity, mean wind speed, etc [5], [6]. The grass temperature is measured using thermometers just above the grass, about 10cm above ground. In [7], the authors used the cloud cover, the atmospheric temperature measured at midnight, and 5-day precipitation and predicted the possibility of frost with 87% of accuracy. Also see Section II, To forecast frost more accurately, several prior studies used microclimate data such as temperatures measured at lower altitude, grass minimum temperature, diurnal, average relative humidity, minimum relative humidity, mean wind speed, etc [5], [6]. The grass temperature is measured using thermometers just above the grass, about 10cm above ground. In [7], the authors used the cloud cover, the atmospheric temperature measured at midnight, and 5-day precipitation and predicted the possibility of frost with 87% of accuracy. And Section II. Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. Solar radiation is calculated as cumulative light intensity per day. See Fig. 1. Sensor Node Kit).
It would have been obvious to one of ordinary skill in the art prior to the effective filing date of the instant application, to modify Moon to explicitly teach wherein the weather observation data include a temperature, humidity, a grass temperature, a wind speed, a soil temperature, an amount of precipitation, and insolation, because several prior studies have used the listed observation data to improve prediction accuracy (Moon p. 1, Col. 2: para. 2).
Regarding Claim 18, Moon does not explicitly teach wherein the frost learning model further comprises a common model used in case that the number of collected weather observation data is equal to or smaller than a predetermined number (Note the remaining recitation after “a common model” is intended use and therefore not given patentable weight).
However, Moon teaches collecting both global and microclimate data (Moon p. 1, Section II. para. 1, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web.).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon to explicitly teach a common model, because while Moon recognizes that having a frost prediction model without microclimate data is less accurate (Moon p. 1, Section I. para. 2, This is because meteorological stations are located farther from the crops whereas the microclimate is measured near the crops, hence there is a difference in altitude, humidity, and other conditions between the global and microclimate measures.), modeling and prediction can still be performed (Moon p. 1, Section I. para. 1, Table I shows a comparison of the global climate collected at the meteorological station and the microclimate collected at four weather stations. There is a noticeable difference between the global and the microclimate data in terms of RMSE (Root Mean Square Error).).
Claim(s) 4, 11, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Moon (as stated above) in view of Lawrence "The Relationship between Relative Humidity and the Dewpoint Temperature in Moist Air: A Simple Conversion and Applications". Bulletin of the American Meteorological Society 86.2 (2005): 225-234. 01 Feb 2005. https://doi.org/10.1175/BAMS-86-2-225).
Regarding Claim 4, Moon further teaches wherein the frost prediction training data include a temperature inversion calculated by using a grass temperature and an ambient air temperature, an amount of precipitation in a predetermined rain time period, an insolation in a predetermined solar time period, an ambient air temperature in a predetermined time period, a temperature difference in a predetermined time period, a wind speed at a predetermined time, a grass temperature in a predetermined time period, a soil temperature in a predetermined time period, a minimum grass temperature in a predetermined time period, and a minimum ambient air temperature in a predetermined time period (Moon p. 1, Section II, Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. […] We start measuring these temperature changes from 5 PM and continue until 11 PM (our forecast time). All collected data has an associated timeframe and time stamp).
Moon does not explicitly teach a dew point generated by using relative humidity and temperature calculation, and a dew condensation in a predetermined time period.
Lawrence teaches a dew point generated by using relative humidity and temperature calculation (Lawrence p. 225 Equation (1)).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application to modify Moon in view of Lawrence to explicitly teach a dew point generated by using relative humidity and temperature calculation, because it is known that dewpoint, temperature, and humidity are interrelated (Lawrence p. 225, Col. 1: para. 1, The relative humidity (RH) and the dewpoint temperature (t d ) are two widely used indicators of the amount of moisture in air.), and therefore the known values can be used to derive the dewpoint.
Moon in view of Lawrence (as stated above) does not explicitly teach a dew condensation in a predetermined time period.
However Moon in view of Lawrence (as stated above) teaches collecting temperature, humidity, and dew point data.
Therefore it would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon in view of Lawrence (as stated above) to explicitly teach a dew condensation in a predetermined time period, by recognizing that water vapor in the air condenses forming dew only when the temperature is less than or equal to the dew point (Note that the limitation under broadest reasonable interpretation may be a binary indication that dew condensation is present.).
Regarding Claim 11, Moon further teaches wherein the frost prediction training data include a temperature inversion calculated by using a grass temperature and an ambient air temperature, an amount of precipitation in a predetermined rain time period, an insolation in a predetermined solar time period, an ambient air temperature in a predetermined time period, a temperature difference in a predetermined time period, a wind speed at a predetermined time, a grass temperature in a predetermined time period, a soil temperature in a predetermined time period, a minimum grass temperature in a predetermined time period, and a minimum ambient air temperature in a predetermined time period (Moon p. 1, Section II, Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. […] We start measuring these temperature changes from 5 PM and continue until 11 PM (our forecast time). All collected data has an associated timeframe and time stamp).
Moon does not explicitly teach a dew point generated by using relative humidity and temperature calculation, and a dew condensation in a predetermined time period.
Lawrence teaches a dew point generated by using relative humidity and temperature calculation (Lawrence p. 225 Equation (1)).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application to modify Moon in view of Lawrence to explicitly teach a dew point generated by using relative humidity and temperature calculation, because it is known that dewpoint, temperature, and humidity are interrelated (Lawrence p. 225, Col. 1: para. 1, The relative humidity (RH) and the dewpoint temperature (t d ) are two widely used indicators of the amount of moisture in air.), and therefore the known values can be used to derive the dewpoint.
Moon in view of Lawrence (as stated above) does not explicitly teach a dew condensation in a predetermined time period.
However Moon in view of Lawrence (as stated above) teaches collecting temperature, humidity, and dew point data.
Therefore it would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon in view of Lawrence (as stated above) to explicitly teach a dew condensation in a predetermined time period, by recognizing that water vapor in the air condenses forming dew only when the temperature is less than or equal to the dew point (Note that the limitation under broadest reasonable interpretation may be a binary indication that dew condensation is present.).
Regarding Claim 20, Moon further teaches wherein the frost prediction training data include a temperature inversion calculated by using a grass temperature and an ambient air temperature, an amount of precipitation in a predetermined rain time period, an insolation in a predetermined solar time period, an ambient air temperature in a predetermined time period, a temperature difference in a predetermined time period, a wind speed at a predetermined time, a grass temperature in a predetermined time period, a soil temperature in a predetermined time period, a minimum grass temperature in a predetermined time period, and a minimum ambient air temperature in a predetermined time period (Moon p. 1, Section II, Currently, all sensor data are collected every minute and summarized as hourly averaged formats in a remote database. […] We start measuring these temperature changes from 5 PM and continue until 11 PM (our forecast time). All collected data has an associated timeframe and time stamp).
Moon does not explicitly teach a dew point generated by using relative humidity and temperature calculation, and a dew condensation in a predetermined time period.
Lawrence teaches a dew point generated by using relative humidity and temperature calculation (Lawrence p. 225 Equation (1)).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application to modify Moon in view of Lawrence to explicitly teach a dew point generated by using relative humidity and temperature calculation, because it is known that dewpoint, temperature, and humidity are interrelated (Lawrence p. 225, Col. 1: para. 1, The relative humidity (RH) and the dewpoint temperature (t d ) are two widely used indicators of the amount of moisture in air.), and therefore the known values can be used to derive the dewpoint.
Moon in view of Lawrence (as stated above) does not explicitly teach a dew condensation in a predetermined time period.
However Moon in view of Lawrence (as stated above) teaches collecting temperature, humidity, and dew point data.
Therefore it would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon in view of Lawrence (as stated above) to explicitly teach a dew condensation in a predetermined time period, by recognizing that water vapor in the air condenses forming dew only when the temperature is less than or equal to the dew point (Note that the limitation under broadest reasonable interpretation may be a binary indication that dew condensation is present.).
Claim(s) 6, 13, 16, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Moon (as stated above) in view of Diedrichs et al. ("Prediction of Frost Events Using Machine Learning and IoT Sensing Devices," in IEEE Internet of Things Journal, vol. 5, no. 6, pp. 4589-4597, Dec. 2018, doi: 10.1109/JIOT.2018.2867333.), hereinafter “Diedrichs”.
Regarding Claim 6, Moon is not relied upon to further teach wherein the weather observation data solves a class imbalance problem of weather observation data by using oversampling of a synthetic minority oversampling technique (SMOTE).
Diedrichs teaches wherein the weather observation data solves a class imbalance problem of weather observation data by using oversampling of a synthetic minority oversampling technique (SMOTE) (Diedrichs p. 4593, Section IV, para. 3, To tackle the scarcity of frost events, we evaluated not only the original datasets, but also datasets with an augmented number of frost events, by using the SMOTE resampling methodology [34], [35]. SMOTE involves a combination of minority class over-sampling, balanced by a majority class under-sampling, resulting in a dataset of the same number of data points. In our experiments, we chose a three-time oversampling of the minority class, that triples the number of frost events.).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon (as stated above) in view of Diedrichs to teach wherein the weather observation data solves a class imbalance problem of weather observation data by using oversampling of a synthetic minority oversampling technique (SMOTE), to improve frost detection sensitivity (Diedrichs p. 4590, Col. 1: para. 4, we propose to use a different approach: we use machine learning algorithms for regression based on Bayesian networks (BNs) and random forest (RF), and for classification based on RF, logistic regression and binary trees, all ran over a balanced training set augmented with new samples produced using the synthetic minority oversampling technique (SMOTE) [6] technique. This technique increases the rate of frost detection (sensitivity).).
Regarding Claim 13, Moon is not relied upon to further teach wherein the weather observation data uses oversampling of a synthetic minority oversampling technique (SMOTE).
Diedrichs teaches wherein the weather observation data uses oversampling of a synthetic minority oversampling technique (SMOTE) (Diedrichs p. 4593, Section IV, para. 3, To tackle the scarcity of frost events, we evaluated not only the original datasets, but also datasets with an augmented number of frost events, by using the SMOTE resampling methodology [34], [35]. SMOTE involves a combination of minority class over-sampling, balanced by a majority class under-sampling, resulting in a dataset of the same number of data points. In our experiments, we chose a three-time oversampling of the minority class, that triples the number of frost events.).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon (as stated above) in view of Diedrichs to teach wherein the weather observation data uses oversampling of a synthetic minority oversampling technique (SMOTE), to improve frost detection sensitivity (Diedrichs p. 4590, Col. 1: para. 4, we propose to use a different approach: we use machine learning algorithms for regression based on Bayesian networks (BNs) and random forest (RF), and for classification based on RF, logistic regression and binary trees, all ran over a balanced training set augmented with new samples produced using the synthetic minority oversampling technique (SMOTE) [6] technique. This technique increases the rate of frost detection (sensitivity).).
Regarding Claim 16, Moon in view of Diedrichs (as stated above) does not explicitly teach wherein the meteorological station is one among a plurality of meteorological stations, and the frost learning model is generated for each of the plurality of meteorological stations.
However, Moon teaches using global climate data and local microclimate data in the frost prediction system (Moon p. 1, Section I. para. 2, Our predictive weather station platform, depicted in Fig. 1, collects microclimate information around crops as well as crop images and global weather information from the web).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application, to modify Moon in view of Diedrichs (as stated above) to explicitly teach wherein the meteorological station is one among a plurality of meteorological stations, and the frost learning model is generated for each of the plurality of meteorological stations, because the microclimate conditions are different in varying localities, and necessary for accurate local conditions, and is well within the inferences and creative steps that a person of ordinary skill in the art would employ (see MPEP 2141.03 I.).
Regarding Claim 17, Moon does not explicitly teach wherein the frost learning model is generated for each of a plurality of predetermined time periods.
Diedrichs teaches wherein the frost learning model is generated for each of a plurality of predetermined time periods (Diedrichs p. 4592, Section IV. A. para. 3 We propose to model a state-based, Gaussian BN which on one hand models both the local distributions of the factorization as normal distributions linked by linear constraints, and on the other represents the state of each variable at discrete time intervals; resulting into a series of time slices, with each indicating the value of each variable at time t.).
It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the instant application to modify Moon in view of Diedrichs to explicitly teach wherein the frost learning model is generated for each of a plurality of predetermined time periods, to supplement that the data used by Moon for training the model is collected every minute and summarized as an average hourly format, because weather varies with time of day and year.
Claim(s) 7 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Moon in view of Lawrence (as stated above), further in view of Scikit-Learn (“3.1. Cross-validation: evaluating estimator performance”. Scikit-learn: Machine Learning in Python. JMLR 12, pp. 2825-2830, 2011, Wayback Machine 03 Dec. 2021), hereinafter “Scikit”.
Regarding Claim 7, Moon in view of Lawrence (as stated above) further teaches wherein a frost prediction model is selected through a verification procedure based on actual data (Moon p. 2, Section II. a), para. 1, We use machine learning toolkits available in GraphLab [8] to train and evaluate five machine learning algorithms. We use 80% data for training and the remaining 20% for testing. The models are tested using actual data).
Moon in view of Lawrence (as stated above) is not relied upon to teach the model is optimized through a grid search and a k-fold cross validation.
Scikit teaches the model is optimized through a grid search (Scikit p. 1, para. 1, The best parameters can be determined by grid search techniques.) and a k-fold cross validation (Scikit p. 1, para. 6, A solution to this problem is a procedure called cross-validation (CV for short). A test set should still be held out for final evaluation, but the vali-dation set is no longer